DNN训练来源不同,可以分为不同的模型,可以看到整合的来源最后预测的效果是最好的(图a)
使用两个模型:BHT(flat)、DARTS-DNN,比较第6天和第0天得到差异可变剪切,再看在exon-intron junction区域的motif的富集(图b)
相关的代码已经放在 Xinglab/DARTS 上面了,可以参考。
DNN predict commad:
➜ bin git:(master) ✗ python Darts_DNN predict -h
usage: Darts_DNN predict [-h] -i INPUT -o OUTPUT [-t {SE,A5SS,A3SS,RI}]
[-e EXPR [EXPR ...]] [-m MODEL]
optional arguments:
-h, --help show this help message and exit
-i INPUT Input feature file (*.h5) or Darts_BHT output (*.txt)
-o OUTPUT Output filename
-t {SE,A5SS,A3SS,RI} Optional, default SE: specify the alternative splicing
event type. SE: skipped exons, A3SS: alternative 3
splice sites, A5SS: alternative 5 splice sites, RI:
retained introns
-e EXPR [EXPR ...] Optional, required if input is Darts_BHT output;
Folder path for Kallisto expression files; e.g '-e
Ctrl_rep1,Ctrl_rep2 KD_rep1,KD_rep2'
-m MODEL Optional, default using current version model in user
home directory: Filepath for a specific model
parameter file
DNN prediction:
# 需要先下载model文件:master.dl.sourceforge.net/project/rna-darts/resources/DNN/v0.1.0/trainedParam/A5SS-trainedParam-EncodeRoadmap.h5
Darts_DNN predict -i darts_bht.flat.txt -e RBP_tpm.txt -o pred.txt -t A5SS
darts_bht.flat.txt:
➜ test_data git:(master) ✗ head darts_bht.flat.txt
ID I1 S1 I2 S2 inc_len skp_len mu.mle delta.mle post_pr
chr1:-:10002681:10002840:10002738:10002840:9996576:9996685 581 0 462 0 155 99 1 0 0
chr1:-:100176361:100176505:100176389:100176505:100174753:100174815 28 0 49 2 126 99 1 -0.0493827160493827 0.248
chr1:-:109556441:109556547:109556462:109556547:109553537:109554340 2 37 0 81 119 99 0.0430341230167355 -0.0430341230167355 0.188
chr1:-:11009680:11009871:11009758:11009871:11007699:11008901 11 2 49 4 176 99 0.755725190839695 0.117542135892979 0.329333333333333
chr1:-:11137386:11137500:11137421:11137500:11136898:11137005 80 750 64 738 133 99 0.0735580941766509 -0.0129207126090368 0
chr1:-:113247674:113247790:113247721:113247790:113246265:113246428 159 1862 127 1958 145 99 0.0550902772187827 -0.0126831240261882 0
chr1:-:1139413:1139616:1139434:1139616:1139223:1139348 980 106 24 2 119 99 0.884944451538756 0.0240073464719553 0.128666666666667
chr1:-:115166127:115166264:115166180:115166264:115165607:115165720 17 0 32 1 151 99 1 -0.0454956312142212 0.287333333333333
chr1:-:11736865:11737005:11736904:11737005:11736102:11736197 22 1532 26 1544 137 99 0.0102705812451076 0.00175172587953396 0
RBP_tpm.txt:
➜ test_data git:(master) ✗ head RBP_tpm.txt
thymus adipose
RPS11 2678.83013 2531.887535
ERAL1 14.350975 13.709394
DDX27 18.2573 14.02368
DEK 32.463558 14.520312
PSMA6 102.332592 77.089475
TRIM56 4.519675 6.14762566667
TRIM71 0.082009 0.0153936666667
UPF2 7.150812 5.23628033333
FARS2 6.332831 7.291382